Google punts sky high storage for devs
Amazon: 'Hey! Don't forget us!'
Google I/O Google has provided limited access to an online storage service built atop its famously distributed online infrastructure.
Known as Google Storage for Developers, the service was unveiled last night with a blog as the company hosts its annual developer conference in San Francisco, California. Using a Representational State Transfer (REST) API, devs can hook their applications into storage that's replicated across multiple top-secret Google data centers in the US.
At the moment, the service is only available to "a limited number" of developers. During this preview period, each dev will received 100GB of data storage and 300GB monthly bandwidth at no charge. You can sign up for the waiting list here.
Google's storage service is akin to Amazon's Simple Storage Service (S3), which dovetails with EC2. Word of Google's announcement leaked onto the web on Tuesday, and by that evening, Amazon had introduced a new S3 option dubbed Reduced Redundancy Storage, or RRS. This lets you save cost by storing non-critical data at lower-levels of redundancy.
"Customers who are using Amazon S3 to share media content that is durably stored elsewhere, or who are storing thumbnails, transcoded media, or other processed data that can be easily reproduced, have told us that they do not always require the level of durability offered by Amazon S3. They have asked us for an even more cost-effective storage solution for this type of data,” Amazon S3 general manager Alyssa Henry said in a statement.
“Reduced Redundancy Storage provides lower redundancy at a lower price, while still giving customers high availability.”
Pricing starts at only $0.10 per gigabyte, and this decreases as you store more data.
Google's service offers multiple authentication methods, SSL support, and controls that let you share access with individuals and groups. Devs can use the service from a web-based interface, or they can use the open-source command-line tool and library GSUtil. ®
Sponsored: Hyper-scale data management